Use informative error message for optional dependency imports (#118)

* Use informative error message for optional dependency imports

* Apply suggestions from code review

Co-authored-by: Anderson Banihirwe <[email protected]>

Author: Max Jones

doc/index.rst

@@ -12,15 +12,41 @@ Installation
 
 Xbatcher can be installed from PyPI as::
 
-    pip install xbatcher
+    python -m pip install xbatcher
 
 Or via Conda as::
 
     conda install -c conda-forge xbatcher
 
 Or from source as::
 
-    pip install git+https://github.com/xarray-contrib/xbatcher.git
+    python -m pip install git+https://github.com/xarray-contrib/xbatcher.git
+
+Optional Dependencies
+~~~~~~~~~~~~~~~~~~~~~
+
+.. note::
+    The required dependencies installed with Xbatcher are `Xarray <https://xarray.dev/>`_,
+    `Dask <https://www.dask.org/>`_, and `NumPy <https://numpy.org/>`_.
+    You will need to separately install `TensorFlow <https://www.tensorflow.org/>`_
+    or `PyTorch <https://pytorch.org/>`_ to use those data loaders or
+    Xarray accessors.
+
+To install Xbatcher and PyTorch via `Conda <https://docs.conda.io/>`_::
+
+    conda install -c conda-forge xbatcher pytorch
+
+Or via PyPI::
+
+    python -m pip install xbatcher[torch]
+
+To install Xbatcher and TensorFlow via `Conda <https://docs.conda.io/>`_::
+
+    conda install -c conda-forge xbatcher tensorflow
+
+Or via PyPI::
+
+    python -m pip install xbatcher[tensorflow]
 
 Basic Usage
 -----------

pyproject.toml

@@ -58,7 +58,7 @@ fallback_version = "999"
 
 [tool.isort]
 profile = 'black'
-known_third_party = ["numpy", "pytest", "setuptools", "sphinx_autosummary_accessors", "tensorflow", "torch", "xarray"]
+known_third_party = ["numpy", "pytest", "setuptools", "sphinx_autosummary_accessors", "torch", "xarray"]
 
 [tool.pytest.ini_options]
 log_cli = true

xbatcher/loaders/keras.py

@@ -1,6 +1,12 @@
 from typing import Any, Callable, Optional, Tuple
 
-import tensorflow as tf
+try:
+    import tensorflow as tf
+except ImportError as exc:
+    raise ImportError(
+        "The Xbatcher TensorFlow Dataset API depends on TensorFlow. Please "
+        "install TensorFlow to proceed."
+    ) from exc
 
 # Notes:
 # This module includes one Keras dataset, which can be provided to model.fit().

xbatcher/loaders/torch.py

@@ -1,6 +1,12 @@
 from typing import Any, Callable, Optional, Tuple
 
-import torch
+try:
+    import torch
+except ImportError as exc:
+    raise ImportError(
+        "The Xbatcher PyTorch Dataset API depends on PyTorch. Please "
+        "install PyTorch to proceed."
+    ) from exc
 
 # Notes:
 # This module includes two PyTorch datasets.